Specific Process Knowledge/Thermal Process/Oxidation

From LabAdviser
Jump to navigation Jump to search

Feedback to this page: click here

Unless otherwise stated, this page is written by DTU Nanolab internal

Oxidation

At DTU Nanolab we have eight furnaces and one RTP (rapid thermal processors) which can be used for thermal oxidation of silicon samples: Boron Drive-in and Pre-dep furnace (A1), Gate Oxide furnace (A2), Phosphorus Drive-in furnace (A3), Anneal-oxide furnace (C1), Anneal-Bond furnace (C3), Al-Anneal furnace (C4), Oxidation 8" furnace (E1), Resist Pyrolysis furnace and RTP Annealsys.

Thermal oxidation can take place either by a dry process or by a wet process, depending on what furnace that is used for the oxidation. The film quality for a dry oxide is better than the film quality for a wet oxide with regards to density and dielectric constant. However, the oxidation rate is slowest for dry oxidation.

  • Dry oxidation is used to grow 5 nm - 300 nm of silicon oxide. Dry oxidation can be done in the A1, A2, A3, C1, C3, C4, E1 and Resist Pyrolysis furnaces.
  • Wet oxidation is used to grow up to ~3 µm of silicon oxide. Wet oxidation can be done in the A1, A3, C1, C3 and E1 furnaces.

Wafers with oxide layers thicker than ~3 µm can normally not be made in the cleanroom and will have to be bought from somewhere else (but check the wafer shop first - there might be some on stock). It is NOT allowed to oxidize the same wafers two times to get a thicker layer than 3 µm without approval.

Thermal oxidation can done at temperatures up to 1050 C - 1150 C, depending on the furnace - and especially the diameter of the quartz tube in the furnace. At these high temperatures, the quartz tube might start to deform, so therefore the oxidation times are restricted:

  • A1, A2, A3, C3 and C4 furnaces: Maximum allowed oxidation time at 1150 C: 8 hours
  • A1, A2, A3, C1, C3, C4 and E1 furnaces: Maximum allowed oxidation time at 1100 C: 23 hours (this will result in ~3 um wet oxide)
  • Resist Pyrolysis furnace: Maximum allowed oxidation time at 1050 C: 3 hours


The standard recipes, quality control limits and results for the Boron Drive-in + Pre-dep furnace (A1) and the Phosphorus Drive-in furnace (A3) can be found here:


The wet oxidation rates for the Anneal-Bond furnace (C1) can be found here:

Standard recipes in the oxidation furnaces

The steps in the standard oxidation recipes in the A-stack furnaces (A1, A2 and A3) and the C-stack furnaces (C1, C3 and C4) can be found here:

Comparison of the oxidation furnaces

Boron Drive-in and Pre-dep furnace (A1)

Gate Oxide furnace (A2)

Phosphorous Drive-in furnace (A3)

Anneal Oxide furnace (C1)

Anneal Bond furnace (C3)

Al-Anneal furnace (C4)

Oxidation 8" furnace (E1)

Resist Pyrolysis furnace (research tool)

RTP Annealsys (research tool)

Generel description
  • Dry and wet oxidation
  • Boron pre-deposition and boron drive-in are also done in the furnace
  • Dry oxidation of e.g. gate oxides layers
  • Dry and wet oxidation
  • Phosphorous drive-in is also done in the furnace
  • Dry and wet oxidation of 100 mm and 150 mm wafers
  • Oxidation and annealing of wafers from the LPCVD furnaces and PECVD4
  • Dry and wet oxidation and annealing of wafers from the wafer bonders and from PECVD4 and PECVD3
  • Dry oxidation of 100 mm wafers and small samples
  • Dry and wet oxidation of 150 mm and 200 mm wafers
  • Dry oxidation of silicon and annealing in N2. But the furnace is mainly being used for pyrolysis of different resists
  • Rapid thermal processing:
    • RTA (annealing)
    • RTO (oxidation)
    • RTN (nitridation)
    • RTH (hydrogenation)
Oxidation method
  • Dry: O2
  • Wet: H2O (torch)
  • Dry: O2
  • Dry: O2
  • Wet: H2O (torch)
  • Dry: O2
  • Wet: H2O (steamer)
  • Dry: O2
  • Wet: H2O (bubbler)
  • Dry: O2
  • Dry: O2
  • Wet: H2O (steamer)
  • Dry: O2
  • Dry: O2
Annealing gas
  • N2
  • N2
  • N2
  • N2
  • N2
  • N2
  • N2
  • N2
  • Ar
  • 5% H2/Ar
Process temperature
  • 800 oC - 1150 oC
  • 800 oC - 1150 oC
  • 800 oC - 1150 oC
  • 800 oC - 1100 oC
  • 800 oC - 1150 oC
  • 400 oC - 1150 oC
  • 800 oC - 1100 oC
  • 25 oC - 1050 oC
  • 700 oC - 1200 oC
Substrate and batch size
  • 1-30 50 mm wafers
  • 1-30 100 mm wafers
  • 1-30 50 mm wafers
  • 1-30 100 mm wafers
  • 1-30 50 mm wafers
  • 1-30 100 mm wafers
  • 1-30 50 mm wafers
  • 1-30 100 mm wafers
  • 1-30 150 mm wafers
  • 1-30 50 mm wafers
  • 1-30 100 mm wafers
  • Small samples on a carrier wafer, horizontal
  • 1-30 50 mm wafers
  • 1-30 100 mm wafers
  • 1-2 150 mm wafers, horizontal, less good uniformity
  • Small samples on a carrier wafer, horizontal
  • 1-50 150 mm wafers
  • 1-50 200 mm wafers
  • 1-30 50 mm wafers
  • 1-30 100 mm wafers
  • 1-30 150 mm wafers
  • Small samples on a carrier wafers, horizontal
  • Single-wafer process
  • Chips on carrier
  • 100 mm or 150 mm wafers
Allowed materials
  • All wafers have to be RCA cleaned, except boron pre-doped wafers from the same furnace
  • All wafers have to be RCA cleaned
  • All wafers have to be RCA cleaned, except phosphorous pre-doped wafers from furnace A4
  • All processed wafers have to be RCA cleaned, except wafers from LPCVD furnaces and PECVD4
  • All processed wafers and samples have to be RCA cleaned, except for wafers from the wafers bonders and from PECVD4 and PECVD3
  • No RCA cleaning required
  • All processed wafers have to be RCA cleaned.
  • Only samples for resist pyrolysis, and all sample materials have to be approved by DTU Nanolab. Samples with metals and III-V materials are NOT allowed
  • Silicon
  • Silicon Nitride
  • Aluminum Oxide


Oxidation curves

Color chart for oxide/nitride thickness

Generic calculator for wet/dry oxide thickness calculation

The following links give an approximate oxide time/thickness based on a general formula:

Deal-Grove parameters

By Kasper Reck-Nielsen February 2015

The following table contains linear and parabolic parameters for use in the Deal-Grove model for thermal oxidation. The parameters are obtained a least squares fit to data available in the furnace logbooks. Information on wafer orientation and doping concentration, which is not available in the logbooks, has not been included in calculating the parameters.

Anneal Oxide Anneal Bond Boron Drive-in Gate Oxide Phosphor Drive-in
Recipe B [µm2/h] B/A [µm/h] RMSE [nm] B [µm2/h] B/A [µm/h] RMSE [nm] B [µm2/h] B/A [µm/h] RMSE [nm] B [µm2/h] B/A [µm/h] RMSE [nm] B [µm2/h] B/A [µm/h] RMSE [nm]
DRY900 0.000408 0.107 Too little data 0.0660 0.272 31 0.000390 0.154 Too little data 0.0028 0.079 4 (limited data) 0.0507 0.884 Too little data
DRY1000 - - - - - - 0.465 0.838 20 - - - 0.641 1.45 41
DRY1050 0.0111 0.526 27 - - - 0.0129 0.330 8 0.022 0.505 3 (limited data) 0.0134 0.362 6
DRY1100 0.020 0.930 10 - - - 0.0212 0.736 23 - - - 0.0313 0.553 14
WET950 - - - - - 0.0716 1.25 12 - - - 0.110 1.17 11
WET1000 0.192 1.54 44 - - 0.192 1.80 29 - - - 0.195 2.49 22
WET1050 0.487 0.965 29 0.477 0.899 Too little data 0.455 1.33 16 - - - 0.448 1.73 12
WET1100 0.580 1.43 8 - - - 0.519 1.186 3 - - - 0.403 9.05 7

Wet Oxidation on <100>

The curves below are based on measurements in our specific furnaces and give more accurate results. We will still recommend to make minimum one test run if the thickness is very important.

  • Wet oxidation

  • A1 Furnace <100>-Si Wet Oxidation

  • A3 Furnace <100>-Si Wet Oxidation

Dry Oxidation on <100> and <111> wafer

  • Dry oxidation

  • Dry oxide on <100> wafer. The y-axis is in Å and the x-axis is in minutes.

  • Dry oxide on <111> wafer. The y-axis is in Å and the x-axis is in minutes.


Breakdown voltage measurements

Breakdown measurements have been done for A1 Boron Drive-in/Pre-dep furnace, A2 Gate Oxide furnace and A3 Phosphorus drive-in furnaces in November 2021. For E1 Oxidation (8") furnace, it has been done in Febuary 2022. The results can be found on this page:

Retrieved from "https://labadviser.nanolab.dtu.dk//index.php?title=Specific_Process_Knowledge/Thermal_Process/Oxidation&oldid=49626"